1. Final Exam Review

2. Data Mining and Data Analytics Techniques Explain the three data analytics techniques we covered in the course Decision Trees, Clustering, and Association Rules What kinds of problems can each solve? Provide a business- oriented example. Make recommendations to a business based on the results of each type of analysis. Explain how data mining differs from OLAP analysis Why would you use this instead of a data cube and a pivot table?

3. Understanding Descriptive Statistics Be able to read and interpret a histogram Negative Skew/Positive Skew Be able to read and interpret sample statistics

4. Sample Statistics VariableMinMaxMean Male01.7 Female01.05.2 Did not reply01.1

5. Decision Tree Analysis

6. Cluster Analysis Scatter Plot What do you look for in a histogram to tell if a variable should be included?

7. Cluster Analysis Cohesion: Similar items being grouped together. Separation: Separate different clusters away from each other. Which cluster has the highest cohesion? Cluster 1. Why? Which cluster has the highest separation? Cluster 3. Why?

8. Cluster Analysis Segment profile Plot Segment 6: Does ORIGINAL have higher values than the data or lower? Higher. Why?

9. Association Rules Be able to read and interpret the output from an association rule analysis Find the strongest (or weakest) rule from a set of output Understand the difference between confidence, lift, and support You should be able to explain the difference between them Can you have high confidence and low lift? You won’t have to compute them, but you should understand how they are computed so you can interpret the statistics

10. What’s the difference between a standard association rule analysis and a sequence analysis? What situations would a sequence analysis make sense? When wouldn’t it? What activity is most likely to be undertaken if website is visited first? Archive. Why? What is a situation where you may have high confidence but not a high lift? Musicstream-> Website. What is happening here?